There is a critical need to develop and evaluate a new signal processing approach that can characterize causal influences between brain regions at the relatively low SNRs typical of noninvasive electroencephalography (EEG) data. The long-term goal of this project is to develop a new set of signal processing tools for study of temporally dynamic network interactions in the cortex from noninvasive measurements of electric or magnetic potentials at the scalp. The objective of this R21 application, the first step toward achieving this long term goal, is to develop and evaluate a novel causal influence framework that integrates the modeling of distributed anatomical cortical sources and multivariate autoregressive (MVAR) temporally dynamic network models. Successful completion of this objective will overcome technical gaps that have severely limited study of cortical networks from EEG data and demonstrate the potential of integrated source/network modeling. The following specific aims will be pursued to achieve this objective: 1) Develop an integrated framework for estimating effective cortical connectivity from scalp measurements;and 2) Evaluate effective connectivity from spontaneous EEG recordings in humans during three behavioral states. The human studies are designed to test the hypothesis that bottom-up information flow is dominant while watching a movie (waking) and top-down information flow is dominant in REM sleep while dreaming. The proposed work is innovative because (i) advanced signal processing methods and EEG source models are employed to infer cortical connectivity directly from scalp recordings, (ii) the experimental data is very likely to exhibit changes in connectivity between states, and (iii) a cross-validation evaluation strategy will provide definitive, quantitative evidence of effectiveness. The outcomes of this project are expected to be: (i) a noninvasive framework for assessing cortical connectivity at realistic SNRs;(ii) a cross-validation methodology for evaluating the quality of cortical network models using experimental data;and (iii) new insights into the nature of consciousness and sleep. This work is significant because development of a tool for reliable, noninvasive assessment of effective connectivity between cortical regions will enable neuroscientists to address many longstanding and new hypotheses about normal brain function, development, and disease. PUBLIC HEALTH RELEVANCE: Contemporary neuroscience has established that human brain function arises from the time-varying neural activity of spatially distributed neural circuits in the cortex. This project develops a new set of signal processing tools for noninvasive interrogation of causal connectivity between different cortical regions and will further understanding of the brain in both health and disease.